Transmeta To Release Next Generation CPU

CodeShark writes: "According to this story at CNN, Transmeta is set to release their new TM6000 microprocessor this afternoon. The chip apparently incorporates some of the functions usually provided by high-performance (and high price!) chip sets. Transmeta is reporting a further reduction in power requirements by 44% and sees the laptop and sub-laptop markert as the primary markets for their new CPU. Intel and AMD claim to be catching up with the Transmeta chips in terms of power requirements, I'd be curious to find out what the real world comparisons might make of those claims ..." If anyone out there is at Microprocessor Forum, please say in comments any further details that are made clear there.

Comments

Re:And yet....

[shayne321]

I have to agree.. I really wanted to like Transmeta. Admittedly at first I fell into the "hey, Linus is behind it so it must be good mindset". Then once they unveiled the code-morphing (or whatever it's called) technology I was really impressed. Wow, what a great idea, I thought - virtualizing the core of the processor and doing optimizations of the x86 instructions on the fly.. Not only should this be faster, but would theoretically allow the chip to run on many different architectures simply by updating the emulation/optimization layer. I thought it was one of the most innovative things I've seen in a while. Somehow they've managed to screw it all up.

First of all, performance has never been there.. They can't even seem to get close to mid-range AMD and intel chips, so they changed position to "well, it's a LOW POWER consumption chip for laptops". Like the previous poster said, even if you half the consumption of the CPU unless you work on the LCD and other components you'll only increase battery life by a few percent. To the average user that's just not worth having to buy a more expensive and unproven chip.

The only other market I could see for them would be in an embedded pc market where a company sold hardware products spanning several architectures and wanted one a single processor they could work with intimately rather than having to learn the quirks of different processors on each architecture they have. Honestly I've racked my brain and can't even think of an example of such a company.. Maybe Cisco? I'm not THAT familiar with their hardware but maybe it spans more than one architecture.

Moral of the story: Just because someone puts out something you enjoy doesn't mean you'll enjoy everything they put out. That's the flawed logic that caused me to actually sit through an entire episode of That's my Bush! (shudder) What a stinking pile of horse-dung that was.

Shayne

Re:Transmeta is a sad joke: Where are the numbers?

[Anonymous Coward]

Their Code Morphing layer is antagonistic towards the benchmarks you mentioned, mainly because it runs lots of different tests that don't allow the code morph software to optimize. On real world tests they'd probably have better results.

I think a fairer comparison would be performance/watts rather than a synthetic bench that doesn't stress how much work you can do. It should also take into consideration the support chips that other traditional CPUs require (it looks like they're building in a bunch of other stuff that you'de need secondary chips for on Intel and AMD).